The present invention relates to a device for reading out adaptation values of self-adapting functions from a motor vehicle control unit at an external readout unit and to a method for reading out adaptation values of self-adapting functions from a motor vehicle control unit.
Self-adapting functions are increasingly used in modern motor vehicle control units. The goal is to adapt the controllers optimally to individual component tolerances, aging effects, or other deviations. A typical area of use is the adaptation of the engine controller in the scope of knock control.
While in the past only a few adaptations were predominantly ascertained, such as offset values that were factors of self-adapting functions, currently more complex memory structures are increasingly used, such as adaptive program maps or adaptive neuronal networks. The current memory structures offer the advantage that various adaptation values may thus be stored as a function of the operating point.
The product of such complex adaptation methods is a parameter set, which describes the program maps content of the adapted program maps and/or the weightings of the adaptive neuronal networks, for example. This parameter set is stored in a nonvolatile memory in the control unit and thus may be used in a following engine cycle.
For repair shops or in the scope of the development of engine controllers, it is often important to read out the current state of the adaptation. In the case of few adaptation values, e.g., for offset values or factors, a direct readout using typical readout units, such as a tester or a unit for development data feedback, is easily possible. German patent document DE 199 48 663 C2 is an example thereof, from which a device for reading out various vehicle data, inter alia, adaptation values, at an external diagnostic unit is known.
A disadvantage of these known devices and methods for reading out the adaptation values is, above all, that, in the event of a large amount of data to be read out, the transmission of all of the data requires a long time.
A further disadvantage results with self-adapting program maps or neuronal networks. The adaptive parameter set describes the value contents at the support points predefined in the controller. In order to be able to assign the adaptation values to the individual operating point, the particular support points must be known in the readout unit, because of which the readout units must be very complex and designed individually. Similar problems result in adapted neuronal networks, in which the parameter set describes the weightings of the adaptive networks.
An object of the present invention is to provide a device improved in regard to the above-mentioned disadvantages and an improved method for reading out the adaptation values of self-adapting functions.
This and other objects and advantages are achieved by a device and a method according to the present invention.
A device according to an exemplary embodiment of the present invention for reading out adaptation values of self-adapting functions from a motor vehicle control unit at an external readout unit includes a motor vehicle control unit, an external readout unit, and a communication interface between the external readout unit and the motor vehicle control unit. The motor vehicle control unit includes a nonvolatile memory for storing adapted parameters or parameter sets of the self-adapting functions, a receiver configured to receive at least one test value set transmitted by the external readout unit, a calculation unit configured to calculate at least one adaptation value as a function of the test value set in the adapted parameters or parameter sets, and a transmitter configured to transmit the calculated adaptation value to the external readout unit.
In a method according to an exemplary embodiment of the present invention corresponding to the device for reading out adaptation values of self-adapting functions of the motor vehicle control unit at the external readout unit, in which adapted parameters or parameter sets of the self-adapting functions are stored in the motor vehicle control unit, the external readout unit transmits a test value set describing a test operating point to the motor vehicle control unit and the motor vehicle control unit calculates at least one adaptation value from the test value set and the adapted parameters or parameter sets and transmits it to the readout unit.
Using the device and/or the method according to the present invention, adaptation values may be read out of the motor vehicle control units using comparatively simple readout units and displayed and/or stored in easily interpretable form. Because the calculation of the adaptation values was already performed directly in the control unit, the method and/or the external readout unit is independent of the adapted function. The outlay for reading out the adaptation values is thus significantly reduced, because in this case no adaptation of the readout units is necessary.
The test value set advantageously describes a test operating point of the self-adapting function. Depending on which operating point for which the adaptations are to be read out, a corresponding operating-point-dependent test value set is transmitted from the readout unit to the motor vehicle control unit.
The test value set may advantageously include one or more operating-point-dependent test values as a function of how many operating points for which the adaptation values are to be read out.
If the test value set includes multiple operating-point-dependent test values, these are advantageously transmitted simultaneously to the motor vehicle control unit and the calculation of the adaptation values is performed sequentially for the individual operating points in the motor vehicle control unit. The ascertained adaptation values are subsequently transmitted back to the readout unit and stored or processed further therein.
At least one of the self-adapting functions advantageously includes an adaptive program map or adaptive neuronal network. The device according to the present invention or the method according to the present invention is especially suitable precisely in self-adapting functions of this type, because in program maps, the adapted parameter set describes the value contents at the support points predefined in the controller. By transmitting an operating-point-dependent test value set, the adaptation values are uniquely assigned to the corresponding test operating point and the particular support points no longer have to be known to the readout unit.
If, in the course of development or in the event of model modifications, the number of support points of the adaptive program maps or the structure of the adaptive neuronal networks changes, the readout unit no longer has to be adapted. This results in a significant reduction in complexity.
The motor vehicle control unit advantageously may be an engine control unit or a transmission control unit, because motor vehicle control units of this type often perform many controls using self-adapting program maps or neuronal networks.
A serial interface may be advantageously used as the communication interface between the motor vehicle control unit and the external readout unit. However, the transmission may also be wireless. The external readout unit may advantageously contain an internal memory and/or be coupled to further devices which perform further analyses using the read out adaptation values.
Other objects, advantages and novel features of the present invention will become apparent from the following detailed description of the invention when considered in conjunction with the accompanying drawings.